As the 2015 Paris climate summit approaches, it is worth noting that this month marks 20 years since the Madrid meeting at which the Intergovernmental Panel on Climate Change (IPCC) declared that “the balance of evidence suggests a discernible human influence on global climate”.

Support for the 1995 IPCC statement came from the physical understanding of heat-trapping properties of greenhouse gases, observations of warming and comparisons of modelled and observed climate-change patterns (climate 'fingerprinting').

Criticism at the time stemmed from the paucity of fingerprint studies, the inadequate treatment of uncertainties, the focus on surface temperature, and poor quantification of natural climate noise and human-caused warming.

One lesson from Madrid is the importance of responding to justifiable criticism. Climate forensics have since identified human-caused temperature fingerprints from the stratosphere to the ocean depths, and in many variables other than temperature. It is routine to assess uncertainties in observed climate data and model simulations. The human-caused warming signal and the noise of natural climate variability are now better quantified. This signal has dominated since the mid-twentieth century.

We have also learned that global scientific understanding can emerge in less than 20 years from the noise of unreason and disinformation, and that one sentence can change the world.